Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/259934948

NewProceduresforControlofIndustrialEffluentsTreatmentProcesses

ARTICLEinINDUSTRIAL&ENGINEERINGCHEMISTRYRESEARCH·JANUARY2014

ImpactFactor:2.59·DOI:10.1021/ie402126d

CITATIONS

4

READS

63

3AUTHORS:

GrzegorzBoczkaj

GdanskUniversityofTechnology

14PUBLICATIONS40CITATIONS

SEEPROFILE

AndrzejPrzyjazny

KetteringUniversity

74PUBLICATIONS1,317CITATIONS

SEEPROFILE

MarianKamińskiGdanskUniversityofTechnology

53PUBLICATIONS418CITATIONS

SEEPROFILE

Availablefrom:GrzegorzBoczkaj

Retrievedon:04February2016

New Procedures for Control of Industrial Effluents TreatmentProcessesGrzegorz Boczkaj,†,* Andrzej Przyjazny,‡ and Marian Kaminski†

†Chemical Faculty, Department of Chemical and Process Engineering, Gdansk University of Technology, G. Narutowicza St. 11/12,80−233 Gdansk, Poland‡Kettering University, 1700 University Avenue, Flint, Michigan 48504, United States

*S Supporting Information

ABSTRACT: This work presents the procedures for monitoring volatile organic compounds during treatment of industrialeffluents. The investigation was carried out for a specific effluentcaustic effluent from bitumen production. The developedprocedures enable more detailed control of the effectiveness of wastewater treatment than standard procedures. Caustic effluentsfrom bitumen production have a complex physicochemical form and consist of an emulsion of an organic phase in a stronglyalkaline aqueous phase. The occurrence of an emulsified organic phase in the aqueous phase of the effluents results in their hightoxicity toward the activated sludge of a refinery wastewater treatment plant as well as strong malodorousness. Interpretation ofanalytical results reveals that the effluents contain over 400 organic compounds among volatile organic compounds (VOCs)alone. Using the developed procedures, 114 of the VOCs were identified in raw postoxidative effluents. The proceduresdescribed in this work allow detailed identification of VOCs as well as the determination of distribution of their concentrationsfor individual classes of chemical compounds. Monitoring changes in content of individual classes of VOCs, including highlymalodorous volatile sulfur compounds, as well as changes in total VOC content, provides more information on the processestaking place during wastewater treatment.

1. INTRODUCTION

Many aspects of environmental protection involve minimiza-tion of the effect of industry on the environment. Thepredominating nature of the effect depends on the type ofindustry, but in the majority of cases the effects are mostlysimilar. The most important effects include emission of airpollutants, emission of pollutants in wastewater and infiltrationthrough the soil, and noise pollution.In each case, identification of major threats and evaluation of

the magnitude of their negative effect are required. Identi-fication of gaseous and liquid pollutants is necessary in order toselect appropriate measures of environmental protection.1−4

Monitoring of transformation of pollutants undergoing treat-ment is also important.5−7 Chromatographic techniques havebeen widely used for screening of pollutants.8−24 Gaschromatography plays a particularly important role in controlof effluent treatment processes in case of occurrence of volatileorganic compounds (VOCs).17,25−29 Electronic noses have alsobeen used in monitoring of the composition of industrialeffluents.30 Monitoring of nonvolatile compounds in effluents isaccomplished by liquid chromatography or flow injectionanalysis. Mass spectrometry is most frequently employed foridentification purposes.31−33 The determination of concen-trations and concentration changes of selected groups ofcontaminants during wastewater treatment can also make use ofother selective detection techniques.34

This work describes detailed identification of volatile organiccompounds and determination of changes in their contentduring treatment of effluents from the production of bitumenusing dynamic headspace and gas chromatography−massspectrometry (DHS-GC-MS). Identification was appended

with the results of investigations employing GC detectorsselective toward sulfur- and nitrogen-containing compounds.Procedures for monitoring changes in content of VOCs in theeffluents resulting from wastewater treatment and based on themeasurement of headspace concentrations of VOCs have alsobeen developed.The investigation was carried out for a specific kind of

effluentsthe alkaline effluent from the production of bitumen.Characteristics of operations and processes resulting information of this type of effluents were described in a previouswork.17 As a result of high loads of pollutants, attempts atpreliminary treatment of the effluent prior to feeding it to thewastewater treatment plant are being undertaken. One of theoperations under consideration is effluent demulsification inorder to separate the organic phase occurring in the effluents.This would result in a substantial decrease in the load ofpollutants, thus lowering the cost of treatment of this kind ofeffluents. The analysis of changes in composition of theheadspace above the effluents resulting from the removal of theorganic phase is presented as an example of proceduresdescribed in this work.

2. MATERIALS AND METHODS2.1. Materials. The following materials were used in this

study: (1) carbon disulfide (GC grade, Fluka Analytical, Neu-Ulm, Germany); (2) standards of volatile organic compounds

Received: July 5, 2013Revised: August 27, 2013Accepted: January 5, 2014Published: January 7, 2014

Article

pubs.acs.org/IECR

© 2014 American Chemical Society 1503 dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−1514

This is an open access article published under an ACS AuthorChoice License, which permitscopying and redistribution of the article or any adaptations for non-commercial purposes.

(Sigma Aldrich, St. Louis, USA); (3) compressed gases helium,nitrogen, air of 5.0 N purity grade (Linde Gaz, Cracow,Poland), and hydrogen of 5.5 N purity grade from a hydrogengenerator (Packard, Downers Grove, USA); (4) Tedlar bags(SKC Inc., Eighty Four, USA); (5) 22 mL headspace vials withcaps equipped with a dpoly(tetrafluoroethylene) (PTFE)-linedsilicone septum (BGB Analytic, Boeckten, Switzerland).2.2. Apparatus. The equipment used is detailed as follows:

(1) a model HP 5890 gas chromatograph with a model HP5972A mass spectrometer (Hewlett-Packard, Wilmington, DE,USA), Chemstation software (Agilent, Santa Clara, CA, USA),and NIST 05 and Wiley 8.0 mass spectra libraries; (2) anAutosystem gas chromatograph with a flame ionization detector(FID) and nitrogen−phosphorus (NPD) detector (Perkin-Elmer, Waltham, Massachusetts, USA), A/C Nelson 900interface (Perkin-Elmer, Waltham, Massachusetts, USA), andTotalchrom software; (3) a model GC 8500 gas chromatographwith FID (Perkin-Elmer, Waltham, Massachusetts, USA), A/CNelson 900 interface (Perkin-Elmer, Waltham, Massachusetts,USA), and Totalchrom software; (4) a model G1901-60502purge and trap concentrator (Hewlett-Packard, Wilmington,DE, USA); (5) GC capillary columns [K1, 60.0 m × 0.25 mm× 0.25 μm DB5 ms (Agilent, Santa Clara, CA, USA); K2, 60.0m × 0.32 mm × 1.0 μm HP1 (Hewlett- Packard, Wilmington,DE, USA); K3, 30.0 m × 0.32 mm × 1.00 μm PE-5 (Perkin-Elmer, Waltham, Massachusetts, USA); K4, empty fused silicacapillary 30.0 m × 0.32 mm (BGB, Boeckten, Switzerland)].2.3. Procedure. 2.3.1. Identification of Volatile Compo-

nents of Postoxidative Effluents by Dynamic Headspace andGas Chromatography−Mass Spectrometry. Samples (5 mL)of the effluents were placed in 10 mL vials. Next, the vials wereclosed with screw caps equipped with a PTFE-lined siliconeseptum. Two fused silica capillaries were introduced throughthe septumone of them fed helium purging the effluentsample while the other one transported the gas with theanalytes to a sorbent trap. The purging process was carried outat 25 °C for 10 min. The sorbent trap was maintained at 30 °C.During thermal desorption the trap was initially heated to 260°C, and the analyte desorption from the trap was performed at270 °C for 4 min. The desorbed analytes were passed through afused silica transfer line heated to 200 °C directly to the gaschromatograph.Separation Conditions. The following separation conditions

were used: (1) column K1; (2) carrier gas of helium at 1.1 cm3/min; (3) injection port temperature of 300 °C (connection ofpurge and trap accessory to GC); (4) temperature program of40 °C (5 min.), ramped at 5 °C/min to 250 °C (15 min.).Mass Spectrometric Detection. The following conditions

for mass spectrometric detection were used: (1) ion sourcetemperature (EI, 70 eV) of 200 °C and GC-MS transfer linetemperature of 310 °C; (2) SCAN mode from mass-to-chargeratio of 34−300 m/z.Three independent determinations were carried out for each

of the investigated sewage samples. Analyte identification wasbased on comparison of mass spectra of the analytes with thosein the NIST and Wiley mass spectra libraries. Peak areas ofidentified analytes were calculated for selected ions character-istic of a particular analyte.2.3.2. Identification of Volatile Sulfur Compounds Using

Static Headspace and Gas Chromatography with PulsedFlame Photometric Detector. Effluent samples (10 mL) werethermostatted in capped 22 mL headspace vials at 80 °C for 30min. Next, 0.5 mL headspace samples were collected using a

gastight syringe heated to 100 °C. The samples wereimmediately introduced into the injection port of gaschromatograph.

Separation Conditions. The following separation conditionswere used: (1) column K2; (2) carrier gas of helium at 1.2 cm3/min; (3) injection port temperature of 300 °C; (4) temperatureprogram of 40 °C (7 min.), ramped at 5 °C/min to 260 °C andramped at 30 °C/min to 300 °C (5 min).

PFPD Operating Parameters. The PFPD operatingparameters were as follows: (1) detector temperature of 300°C and gas pressures for hydrogen of 21.0 psig, for air of 12.6psig, and for makeup gas (air) of 14.8 psig; (2) signalacquisition parameters for photomultiplier tube voltage of 600V, for a pulse rate of 3.57 Hz, and for the sulfurchemiluminescence gate width of 6−24 ms.A detailed description of the procedure and method

validation is provided in ref 17.2.3.3. Identification of Volatile Nitrogen Compounds by

Static Headspace and Gas Chromatography with Nitrogen−Phosphorus Detector . Effluent samples (10 mL) werethermostatted in capped 22 mL headspace vials at 80 °C for30 min. Next, 0.5 mL headspace samples were collected using agastight syringe heated to 100 °C. The samples wereimmediately introduced into the injection port of the gaschromatograph.

Separation Conditions. The following separation conditionswere used: (1) column K3; (2) carrier gas of helium at 2.0 cm3/min; (3) injection port temperature of 300 °C; (4) temperatureprogram of 45 °C (2 min), ramped at 10 °C/min to 285 °C (15min).

Operating Parameters of NPD. The operating parametersof the nitrogen-phosphorus detector (NPD) were as follows:detector temperature of 300 °C and gas flow rates for hydrogenof 3.5 cm3/min and for air of 85 cm3/min.Three independent determinations were performed for each

analysis. Analyte identification was based on comparison ofretention times with those for a standard mixture.Concentrations of volatile nitrogen compounds in the

headspace were calculated using response factors for theNPD detector.The limit of detection (LOD) was calculated from eq 1:

= SN

LOD 3(1)

where S is the analyte signal (pA) and N is noise near theanalyte retention time (pA). The limit of quantitation (LOQ)was calculated from eq 2:

=LOQ 2LOD (2)

The response factor was calculated from eq 3:

= RAmi

i

if,

(3)

where Rf,i is the response factor of analyte i (μV·s/ng), Ai is theaverage value (n = 3) of the peak area of analyte i (μV·s), andmi is the mass (ng) of analyte i introduced into the GC.The response factors per nanogram of analyte as well as their

limits of detection and quantitation determined for four volatilenitrogen compounds are compiled in Table 1.In procedures based on headspace analysis, there is a strong

effect of the matrix composition on the solution−headspaceequilibrium. In the case of industrial effluents of a variable

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141504

composition and pH (10.0−11.5), the ratio of partitioncoefficients of individual compounds with respect to a specificinternal standard will not be constant. For this reason, theinternal standard method was not used in this work.The proposed procedure is based on comparison of

concentrations determined directly in the headspace for raweffluent and that after treatment.2.3.4. Determination of Total Content of Volatile Organic

Compounds by Static Headspace and Gas Chromatographywith Flame Ionization Detector. Effluent samples (10 mL)were thermostatted in capped 22 mL headspace vials at 50 °Cfor 30 min. Next, 0.5 mL headspace samples were collectedusing a gastight syringe heated to 80 °C. The samples wereimmediately introduced into the injection port of the gaschromatograph.Separating Conditions. The following separation conditions

were used: (1) column K4; (2) injection port temperature of300 °C and split injection mode (5:1); (3) oven temperature of275 °C; (4) carrier gas of helium at 5.0 cm3/min.Operating Parameters of FID. The FID operating

parameters were as follows: detector temperature of 300 °C,gas flow rates for hydrogen of −40 cm3/min and for air of −450cm3/min.The concentrations of volatile organic compounds in the

headspace were determined using the external standard method(five-point calibration curve). Methane in air was used as thestandard gas. In order to test the accuracy of determinationsmaking use of a universal calibration with methane todetermine total VOC content, the results were comparedwith the ones for a standard mixture containing eight chemicalcompounds.Preparation of Standard Mixtures. Standard gaseous

mixtures were prepared by injecting liquid standard compoundsinto a Tedlar bag using a microsyringe. Eight compounds withvarying polarities, boiling points, and structures were selected:2-propanethiol, hexane, 2-propanol, cyclohexane, 2-butanone,toluene, 2-methylpyridine, and acetonitrile. Tedlar bags werefilled with synthetic air introduced by means of a gastightsyringe from a compressed air tank through a septum installedon a gas line. The composition of the primary standard mixturewas calculated from the amounts of standards introduced,assuming their complete evaporation. Other standard mixtures

were prepared by successive injections of known volumes of amore concentrated standard mixture into another Tedlar bagfilled with synthetic air. In this way, gas standards with totalconcentrations of 10, 100, and 1000 ppm were prepared. Priorto usage, Tedlar bags were cleaned five times by repeating acleaning cycle consisting of filling the bag with high puritynitrogen, leaving it for 30 min, and emptying it using a vacuumpump. Prior to preparation of standard gaseous mixtures, theTedlar bag background was checked by carrying out threeanalyses of air using GC-FID.Analytical characteristics of the procedure for the determi-

nation of total VOCs are discussed in the following sections.Linearity. Linearity of the procedure was examined in the

total VOC content range described above using the preparedstandard mixture and relative to the calibration mixture(methane in air), against which quantitative analysis wascarried out. Linearity of the detector response in theinvestigated range was estimated from the regression coefficient(r), which should be equal to or close to one. For theinvestigated procedure, this assumption was met in the range1−104 ppm.Limit of detection and quantitation of the GC-FID

procedure for determination of the total VOC content wascalculated as follows.The detection limit (LOD) was calculated from eq 4:

=s

bLOD

3.3 a(4)

where LOD is the limit of detection, sa is the standard deviationof the intercept of the calibration curve, and b is the slope of thecalibration curve.The limit of quantitation (LOQ) was calculated from eq 2.Table 2 lists numerical values of the investigated validation

parameters.A comparison of the characteristics of FID responses for the

standard VOC mixture and for methane reveals that the totalVOC content can be calculated from the calibration curveobtained for methane in air.

2.3.5. Effluent Demulsification. Effluent of 150 mL volumewas mixed in a separatory funnel with 15 mL of a 0.2% (w/w)solution of industrial demulsifier for refinery effluents. After 5min of manual shaking, the emulsion was left to stand toaccomplish phase separation. The lower fractionthe effluentdevoid of organic phasewas transferred to tightly closed vials.Following demulsification operation, the postoxidative effluenthad the appearance of a clear aqueous phase. Raw anddemulsified effluents were stored in the vials at 4 °C untilanalysis. The determinations were carried out in time notexceeding 24 h after demulsification.

2.3.6. Determination of Percent Reduction in VOCContent by Effluent Demulsification Using DHS-GC-MS.Individual chromatographic peaks were identified by compar-ison of their mass spectra with those in the NIST and Wileylibraries. Two characteristic mass-to-charge ratio values wereselected for each compound. This formed the basis for VOC

Table 1. Compilation of Experimental Response Factors andLimits of Detection and Quantification for Four StandardVOCs

compoundRf,i

a (μV·s/ng) RSD (%)

LODb (pg(ppm))

LOQc (pg(ppm))

acetonitrile 758918 3.4 48 (0.06) 96 (0.11)pyridine 987811 3.1 32 (0.02) 64 (0.03)2-methylpyridine 688812 3.3 51 (0.03) 102 (0.05)2,4-dimethylpyridine

521700 3.1 64 (0.03) 128 (0.06)

aRf,i = response factor. bLOD = detection limit. cLOQ = quantificationlimit.

Table 2. Comparison of Regression Parameters of Calibration Curves

mixture slope intercept R2a LOD (mg/m3) LOQ (mg/m3) LOQ (ppm)

methane in air 154.34 ± 0.78 0.134 ± 0.024 0.999 0.6 1.2 1.8mixture of VOCs in air 150.12 ± 0.79 0.70 ± 0.23 0.999 1.1 2.2 0.7

aR2 = determination coefficient.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141505

identification. Integration of chromatographic peaks was basedon the detector signal counted only for the selected ions foreach compound. The results were compared for theinvestigated effluent before and after removal of the organicphase.The headspace concentrations determined for volatile sulfur

compounds (VSCs), volatile nitrogen compounds (VNCs), andtotal VOCs are reported in units of mass per headspace volumeand also converted to mole fractions in the headspace using eq5

=⎡⎣

CC

M/ppm

24.45 /(mg/m )]3

(5)

where M is the molar mass of a compound (g/mol).Conversion to parts per billion takes place after substitutionin eq 5 concentration expressed in (μg/m3).

3. RESULTS AND DISCUSSIONCaustic postoxidative effluents are a complex mixture consistingof a strongly basic aqueous phase containing a highconcentration of sulfides (from absorption of hydrogen sulfide),volatile organic compounds, and a condensate of oil mist andbitumen microdrops carried from the oxidation reactor by astream of vapors. Microscopic examination of the effluentmorphology revealed that the effluent is a stable emulsion ofthe organic phase in the aqueous phase. Formation of theemulsion can likely be attributed to the presence in the effluentof surfactants formed during the oxidation process, i.e., long-

chain aldehydes or carboxylic acids. Postoxidative effluents(alkaline absorbate), prior to being fed to a wastewatertreatment plant, are recycled in a scrubber in which wastegases are washed from the bitumen oxidation installation.Pumping the effluent intensifies the process of emulsification ofthe organic phase in the aqueous phase. Depending on theintensity of the oxidation process, fresh batches of lye are addedto the cycled absorbate, and excess effluent is removed from theinstallation. After preliminary isolation of the oil condensate ina plate separator, the effluents are fed to a wastewater treatmentplant. The effluents undergo several stages of treatment,including physical separation of the oil phase, flocculation,biological treatment, and clarification. As a result of strongemulsification of the effluent, classical methods of wastewatertreatment are not sufficiently effective for postoxidativeeffluents. Consequntly, biological treatment is performed onthe effluent with excessively high load of organic pollutants.Specifically, high biotoxicity and malodorousness of theeffluents call for their preliminary treatment prior tointroduction into a wastewater treatment plant. This, in turn,requires the development of an effective technology of effluentpretreatment, ensuring a significant reduction in theirmalodorousness and biotoxicity.This work describes analytical procedures allowing more

effective control of the pretreatment effectiveness than thestandard procedures permit. In many cases, process control ofthe effectiveness of effluent pretreatment is accomplished bydetermining standard water quality parameters, such as

Figure 1. DHS-GC-MS chromatogram of raw postoxidative effluent, enlarged between 5 and 15 min. Identified compounds: (1) propanone; (2) 2-methylpropanal; (3) 2,5-dihydrofuran; (4) butanal; (5) 2-butanone; (6) 2-butanol; (7) 3-methylbutanal; (8) 3-methylbutanone; (9) 2-methylbutanal; (10) benzene; (11) 2-methylhexane; (12) 2-pentanone; (13) 2-methylpentenal; (14) 2-ethoxy-1-propanol; (15) heptane; (16)methylcyclohexane; (17) methyl isobutyl ketone; (18) dimethyl disulfide; (19) 2-methyl-3-pentanone; (20) 3-ethoxy-2-methylpropene; (21) 2-isopropyl-2-methyloxirane; (22) toluene; (23) 2-methylheptane; (24) 3-hexanone; (25) 2-hexanone; (26) 2,5-dimethylhexane; (27) cis-1-butyl-2-methyl-cyclopropane; (28) 3-methyl-2-heptanol; (29) octane; (30) 2-methylhexanal; (31) 3,5-dimethylcyclohexane; (32) ethyl methyl sulfide; (33)2,6-dimethylheptane; (34) 5-methyl-2-hexanone; (35) 1,1,3-trimethylcyclohexane; (36) 2-heptanone; (37) 2-nonene; (38) ethylbenzene; (39) 1-octene; (40) p-xylene; (41) 3-methyloctane; (42) 3-heptanone; (43) o-xylene; (44) 2-ethyl-2-pentenal; (45) nonane.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141506

biological oxygen demand (BOD) and chemical oxygendemand (COD). However, BOD and COD are not goodmeasures of the effectiveness of effluent treatment aimed atreduction of malodorousness, i.e., reduction of specific groupsof VOCs as well as their total content.No works characterizing the composition of caustic effluents

from the bitumen production have been published thus far.Examination of the effluent composition provides additionalinformation on the components formed during bitumenoxidation by thermal cracking and oxidation of the resultingchemical compounds. The analysis of raw composition of theeffluents allows estimating environmental hazards associatedwith possible emission of volatile components from theeffluents during treatment processes carried out in open basins.This pertains especially to biological stages of wastewatertreatment (activated sludge process).The effluent from the bitumen production studied in this

project were collected from industrial scale installation for thebitumen oxidation behind the pump transporting the effluent toa corrugated plate separator, used for the removal of organicphase condensate from the effluent stream. As was mentionedearlier, this kind of separation of organic phase is ineffective forpostoxidative effluent, which contains the organic phase notonly in the form of condensate, but also in the form ofemulsion in the aqueous phase of the effluent. Under industrial

conditions, demulsification as a step of pretreatment ofpostoxidative effluent should precede separation by acorrugated plate separator. This way, both the condensate aswell as the organic phase separated from the effluent throughdemulsification would be removed. The results of largelaboratory-scale investigations on demulsification of postox-idative effluents will be the subject of future papers.Demulsified effluent devoid of organic phase can be pumpedto a wastewater treatment plant and subjected to biologicaltreatment. If demulsified effluent has to be treated further toremove the pollution load, it can be subjected to otherpretreatment stages, e.g., through oxidation.

3.1. Identification of VOCs in Postoxidative EffluentsUsing Dynamic Headspace Followed by Gas Chroma-tography−Mass Spectrometry. Identification of VOCs inpostoxidative effluents was carried out by using dynamicheadspace coupled with gas chromatography−mass spectrom-etry (DHS-GC-MS). The effectiveness of analyte enrichmentby purge and trap depends mostly on analyte volatility.Inspection of n-alkanes identified by GC-MS revealed that theheaviest n-alkane released from the effluents by purge and trapis n-tetradecane (n-C14; boiling point, 254 °C).In the case of using dynamic headspace technique, a negative

effect of water vapor on the sorption capacity of the trap is welldocumented, even when using hydrophobic sorbents, such as

Figure 2. DHS-GC-MS chromatogram of raw postoxidative effluent, enlarged between 15 and 20 min. Identified compounds: (46) 1,2-dimethyl-3-(1-methylethyl)cyclopropane; (47) 4-methoxy-4-methyl-1,3-pentadiene; (48) 2-methyl-2-oxazoline; (49) cyclohexanone; (50) 6-methyl-2-heptanone; (51) benzaldehyde; (52) pentylcyclopentane; (53) propylbenzene; (54) 4-methylnonane; (55) 1,1,2,3-tetramethylcyclohexane; (56)2,6-dimethyl-2,6-octadiene; (57) 5-methyl-3-heptanone; (58) 1-ethyl-4-methylbenzene; (59) cyclodecane; (60) 1,2,4-trimethylbenzene; (61)decane; (62) 1,2,4-trimethylbenzene; (63) 2-methylcycloheptanone; (64) acetophenone; (65) 2-methylbenzaldehyde; (66) 3-methylbenzaldehyde;(67) 2-nonanone; (68) 1-ethyl-2,4-dimethylbenzene; (69) undecane; (70) 1,2,4,5-tetramethylbenzene; (71) 1,2,4,6-tetramethylbenzene; (72) 4-methylundecane; (73) 1-ethyl-3,5-dimethylbenzene; (74) 1-(4-methylphenyl)ethanone; (75) 2-butyl-1-octanol.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141507

Tenax TA. The use of elevated temperatures for purging theanalytes from the aqueous phase is disadvantageous. It wouldresult in a shorter time of purging but at the expense ofproblems with excessive amounts of water vapor introducedwith the purge gas. For this reason, in order to utilize thedeveloped procedure in routine analyses, it was decided topurge the analytes at ambient temperature, i.e., 25 °C. Thesorption trap temperature during the enrichment step was set at30 °C. This temperature ensures appropriate capacity of thesorption trap while at the same time enables rapid coolingfollowing the thermal desorption step solely by using the built-in fan in the DHS accessory. The use of a gaseous cooling agent(vapor of liquid nitrogen, CO2) to cool the trap to lowertemperatures during routine analyses is tedious and impractical.The use of DHS-GC-MS resulted in identification of 87

VOCs. Examples of DHS-GC-MS chromatograms of post-oxidative effluents are shown in Figures 1−3 and S1(Supporting Information).The results of investigation of the headspace of postoxidative

effluents demonstrated the presence of a large number ofvolatile organic compounds. Eighty-seven VOCs were identifiedin raw postoxidative effluents using DHS-GC-MS. Nocarboxylic acids were found in the headspace above theeffluents due to strongly alkaline pH of the effluents. Underthese conditions, physicochemical equilibrium solution−head-space for acidic compounds is strongly shifted toward solutionand ionized form. Consequently, acidic compounds are notreleased to the headspace, which is advantageous, since it

prevents malodorous acidic sulfur compounds or carboxylicacids from being released to the environment.Inspection of DHS-GC-MS chromatograms in Figures 1−3

reveals the presence over 300 chromatographic peaks ofanalytes released from raw postoxidative effluents. Identificationof a larger number of compounds would require the use of ahigh performance capillary column with a length exceeding 100m, preliminary fractionation of sample components prior tochromatographic separation, or the use of multidimensionalseparation,24,35 e.g., coupling liquid and gas chromatography orusing two-dimensional gas chromatography (GC×GC). Ana-lyte identification based on DHS-GC-MS is supplemented bythe results of investigations using highly selective nitrogen andsulfur detectors (Tables 4 and 5).Due to the presence in the effluents of emulsified organic

phase amounting to several percent of the effluent mass, theeffluents are highly toxic to the activated sludge. The effluentsare also characterized by high contaminant load (the averageCOD on the order of 12,000 mg of O2/dm

3) and highmalodorousness. The analysis of changes in content ofindividual volatile organic compounds in the effluents resultingfrom the removal of emulsified organic phase reveals that asignificant reduction in the load of pollutants is possible. Theuse of demulsification during preliminary effluent treatment isadvantageous due to low process and investment costs ofimplementation of such a technology and high effectiveness ofreduction of pollutant content. The percent reduction inpollutant load resulting from the removal of emulsified organic

Figure 3. DHS-GC-MS chromatogram of raw postoxidative effluent, enlarged from 20 min up. Identified compounds: (76) dodecane; (77) 2,6-dimethylundecane; (78) 1,3-dimethyl-5-(1-methylethyl)benzene; (79) 1-ethyl-2,4,5-trimethylbenzene; (80) 2-butyl-1,1,3-trimethylcyclohexane; (81)hexylbenzene; (82) tridecane; (83) cyclotridecane; (84) 2,4,5-trimethylbenzaldehyde; (85) 2,4,6-trimethylbenzaldehyde; (86) 2,3,5,6-tetramethyldecane; (87) tetradecane.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141508

Table 3. Comparison of Removal Efficiency of Individual Classes of VOCs Resulting from Removal of Emulsified Organic Phase

raw effluent demulsified effluent

compound A1a (n = 3), sb (pA*s) A1

a (n = 3) sb (pA*s) reduction (%)

Aldehydes2-methylpropanal 7674 317 6224 227 18.9butanal 79053 3062 56928 2262 28.02-methylbutanal 8729 395 6985 322 20.03-methylbutanal 8607 334 6985 266 18.82-methylpentanal 39695 1558 31393 1091 20.9hexanal 11535 477 10399 478 9.82-methyl-2-pentanal 23533 1073 12619 610 46.42-ethyl-2-pentenal 15347 619 6178 235 59.7benzaldehyde 12713 498 4159 189 67.32-methylbenzaldehyde 19962 839 7375 301 63.14-methylbenzaldehyde 5024 185 3275 109 34.82,4,5-trimethylbenzaldehyde 5911 217 1587 62 73.22,4,6-trimethylbenzaldehyde 3042 124 1936 74 36.4

Alkanes and Alkenes2-methylhexane 5127 202 not found ≫99,9heptane 8076 333 not found ≫99,92-methylheptane 3674 151 not found ≫99,92,5-dimethylhexane 4973 186 not found ≫99,9octane 11572 426 not found ≫99,92,6-dimethylheptane 21247 875 not found ≫99,93-ethylheptane 2273 92 not found ≫99,93-methyloctane 20469 782 not found ≫99,9nonane 25150 1044 not found ≫99,94-methylnonane 4109 176 not found ≫99,9decane 34195 1340 501 25 98.5undecane 29809 1112 1099 51 96.34-methylundecane 11489 474 not found - ≫99,9dodecane 19743 909 not found ≫99,92,6-dimethylundecane 2179 74 not found ≫99,9tridecane 27691 930 1566 68 94.32,3,5-trimethyldecane 1803 80 776 33 57.0tetradecane 2016 79 965 42 52.11-octene 9143 311 3129 125 65.84-methoxy-4-methyl-1,2-pentadiene 15347 674 6198 244 59.62,6-dimethyl-2,6-octadiene 8741 431 not found ≫99,9

Alcohols2-butanol 78349 3456 57019 2368 27.22-ethoxy-1-propanol 1793 67 1470 49 18.02-hexanol 590 22 480 21 18.62-butyl-1-octanol 16693 675 1536 74 90.8

Aromatic Compoundsbenzene 56787 2772 22113 1032 61.1toluene 71637 2389 15980 630 77.7ethylbenzene 85887 3395 6181 288 92.8p-xylene 85887 4696 6180 264 92.8o-xylene 22134 873 1979 94 91.1propylbenzene 13082 533 461 24 96.51-methyl-3-propylbenzene 18798 765 686 37 96.41,2,4-trimethylbenzene 34957 1425 1719 77 95.11,2,5-trimethylbenzene 32927 1150 2526 108 92.31-ethyl-2,4-dimethylbenzene 13478 504 not found ≫99,91,2,4,5,-tetramethylbenzene 21102 899 1774 88 91.61,2,4,6-tetramethylbenzene 16472 611 not found ≫99,91-ethyl-3,5-dimethylbenzene 65434 2464 10901 415 83.31,3-dimethyl-5-(1-methylethyl)benzene 43800 1695 6205 313 85.81-ethyl-2,4,5-trimethylbenzene 41247 1830 not found ≫99,9hexylbenzene 26725 1208 726 43 97.3

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141509

phase determined by the developed DHS-GC-MS procedure islisted in Table 3. Inspection of Table 3 reveals that the contentof all classes of organic compounds undergoes a significantreduction. The degree of reduction in content increases withhydrophobicity of chemical compounds present in the effluents.The highest effectiveness of reduction in content was achievedfor alkanes, cycloalkanes, and aromatic compounds. In theremaining groups of organic compounds the effectiveness ofreduction in content increases with the length of alkyl chain.3.2. Identification and Distribution of VSC Concen-

trations in the Effluent Headspace. Mirror image SHS-GC-PFPD chromatograms of the headspace of raw postoxidativeeffluent and the effluent from which organic phase wasremoved are shown in Figure 4.The concentrations of identified and approximate concen-

trations of unidentified volatile sulfur compounds in theheadspace of raw postoxidative effluent and the demulsifiedeffluent are compared in Table 4. Twenty-four volatile sulfurcompounds were identified in the headspace and thechromatogram contained over 50 other VSCs. The compoundsare formed during thermal cracking of high molecular weightsulfur compounds occurring in the vacuum residue undergoingoxidation. A fraction of VSCs is oxidized during bitumenproduction. Their presence in the effluents results mainly from

the high volatility of VSCs which, under conditions of bitumenoxidation, are mostly removed from the bitumen mass with hotair and steam. Washing with an aqueous solution of sodiumhydroxide assures nearly quantitative absorption of acidic VSCs.Their retention is based on chemical absorption with theformation of respective organic sulfides. Their release from theeffluent can only take place if effluent pH is lowered.17 Underconditions used for the SHS-GC-PFPD analyses, the head-space−alkaline solution equilibrium is strongly shifted towardthe ionized form.The data shown in Table 4 reveal that a substantial reduction

in content of some VSCs is possible during the firstpostoxidative effluent pretreatment stage, i.e., demulsificationas a way of removing the organic phase. A decrease in VSCcontent in the effluent is desirable due to their malodorousness,which can be experienced during their release to theatmosphere at successive stages of treatment in a wastewatertreatment plant. These stages are generally carried out in opentanks, in which separation of the organic phase is accomplishedthrough flotation, flocculation, and biological treatment.

3.3. Identification and Distribution of VNC Concen-trations in the Effluent Headspace. Static headspace andgas chromatography with nitrogen−phosphorus detector (SHS-GC-NPD) chromatograms of raw postoxidative effluents and

Table 3. continued

raw effluent demulsified effluent

compound A1a (n = 3), sb (pA*s) A1

a (n = 3) sb (pA*s) reduction (%)

Cycloalkanesmethylcyclohexane 5419 216 not found ≫99,93,5-dimethylcyclohexane 7569 319 not found ≫99,91,1,3-trimethylcyclohexane 11476 548 not found ≫99,91,2-dimethyl-3-(1-methylethyl)cyclopropane 8142 366 not found ≫99,91,2,3-trimethylcyclohexane 3844 163 1535 76 60.11,1,2,3-tetramethylcyclohexane 14721 594 not found ≫99,9cyclodecane 7499 336 not found ≫99,92-butyl-1,1,3-trimethylcyclohexane 47319 1499 4614 280 90.2cyclotridecane 6487 295 857 29 86.8

Ethers2,5-dihydrofuran 2485 105 1578 74 36.5allyl methallyl ether 1522 63 not found ≫99,92-methyl-2-oxazoline 7596 328 4422 192 41.82-isopropyl-2-methyloxirane 2080 99 2057 102 1.13-ethoxy-2-methylpropene 3148 134 2990 136 5.0

Ketones2-butanone 79053 3132 57019 2375 27.93-methylbutanone 8729 417 6985 345 20.02-pentanone 408089 15422 297377 11485 27.1methyl isobutyl ketone 11146 409 3247 139 70.92-methyl-3-pentanone 3148 155 2990 149 5.03-hexanone 10777 415 10399 403 3.55-methyl-2-hexanone 4088 167 1586 74 61.22-heptanone 1296 66 760 39 41.43-heptanone 7342 315 5064 216 31.0cyclohexanone 6971 316 3690 171 47.15-methyl-3-heptanone 4714 190 2336 111 50.42-methylcycloheptanone 7369 319 826 51 88.8acetophenone 19924 740 7265 276 63.52-nonanone 2841 115 1287 65 54.71-(4-methylphenyl)ethanone 24748 972 2335 97 90.66-methyl-2-heptanone 10382 494 4936 235 52.5

aA1 = average peak area. bs = standard deviation.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141510

Figure 4.Mirror image SHS-GC-PFPD chromatograms of raw postoxidative effluent (blue) and the effluent from which organic phase was removed(red). Identified compounds: (1) hydrogen sulfide; (2) ethanethiol; (3) carbon disulfide; (4) 2-propanethiol; (5) 1- propanethiol; (6) thiophene;(7) dimethyl disulfide; (8) 3-methyl-1-butanethiol; (9) 3-methylthiophene; (10) 2-ethylthiophene; (11) dipropyl sulfide; (12) diethyl disulfide; (13)1-hexanethiol; (14) 1,3-propanedithiol; (15) thiophenol; (16) 1,4-butanedithiol; (17) di-tert-butyl sulfide; (18) 1-heptanethiol; (19) dibutyl sulfide;(20) dipropyl disulfide; (21) benzothiophene; (22) 1-nonanethiol; (23) dihexyl sulfide; (24) 1-decanethiol.

Table 4. Comparison of Headspace Concentrations of Volatile Sulfur Compounds of Raw Postoxidative Effluent andDemulsified Effluent

raw postoxidative effluent demulsified effluent

compound

C1a

(μg/m3;n = 3)

sb

(μg/m3)

recalcd headspace concnexpressed as mole fraction (ppb)

C1a

(μg/m3;n = 3)

sb

(μg/m3)recalcd headspace concn

expressed as mole fraction (ppb) reduction (%)

hydrogen sulfide 94.8 4.1 71.8 89.2 3.6 67.6 5.9ethanethiol 385.0 13.3 291.7 45.1 1.7 34.1 88.3carbon disulfide 94.0 3.6 71.2 89.0 3.7 67.4 5.32-propanethiol 171.4 5.2 129.9 33.1 1.3 25.1 80.71-propanethiol 93.6 3.8 70.9 <LOQ 94.3c

thiophene 90.1 3.8 68.3 28.3 1.1 21.5 68.6dimethyl disulfide 695.5 23.0 527.1 34.9 1.5 26.5 95.03-methyl-1-butanethiol 107.1 4.2 81.2 71.0 2.8 53.8 33.73-methylthiophene 23.6 1.2 17.8 23.2 0.9 17.6 1.52-ethylthiophene 100.7 4.3 76.3 99.6 4.5 75.5 1.1dipropyl sulfide 153.1 5.9 116.0 <LOQ 92.5c

diethyl disulfide 39.6 1.5 30.0 25.4 1.0 19.3 35.91-hexanethiol 95.9 4.6 72.7 37.7 1.6 28.6 60.71,3-propanedithiol 25.9 1.3 19.7 <LOQ 66.6c

thiophenol 88.5 3.8 67.1 69.1 1.9 52.4 21.91,4-butanedithiol 126.6 5.0 95.9 24.3 1.0 18.4 80.8di-tert-butyl sulfide 188.1 7.5 142.5 101.7 5.3 77.1 45.91-heptanethiol 69.0 3.1 52.3 47.9 2.2 36.3 30.6dibutyl sulfide 32.4 1.4 24.5 31.1 1.4 23.5 4.0dipropyl disulfide 60.9 2.3 46.2 60.1 2.4 45.5 1.4benzotiophene 60.1 2.5 45.5 59.4 1.9 45.0 1.01-nonanethiol 50.2 1.8 38.0 <LOQ 82.4c

dihexyl sulfide 66.4 3.2 50.3 53.2 2.0 40.3 20.01-decanothiol 17.7 1.0 13.4 <LOQ 65.6c

sum of identified VSCs 2930 111 2221 1064 42 806 63.7sum of unidentifiedVSCs

5813 273 4405 4645 202 3520 25.0

aCi = average headspace concentration. bs = standard deviation. cDifference calculated on the basis of the value in the interval LOD < X < LOQ.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141511

demulsified effluents are shown in Figures 5 and 6, respectively.

Similarly to volatile sulfur compounds, a much richer

composition of the headspace compared to bitumen samples

was observed.

A comparison of headspace concentrations of volatilenitrogen compounds of raw postoxidative effluent anddemulsified effluent is shown in Table 5.The removal of organic phase from the effluents results in a

substantial decrease in concentration of volatile nitrogencompounds in the effluents. The reduction in content of

Figure 5. SHS-GC-NPD chromatogram of raw postoxidative effluent. Identified compounds: (1) pyridine; (2) 2-methylpyridine; (3) 2,4-dimethylpyridine.

Figure 6. SHS-GC-NPD chromatogram of postoxidative effluent following demulsification. Compound identification as in Figure 5.

Table 5. Comparison of Headspace Concentrations of Volatile Nitrogen Compounds of Raw Postoxidative Effluent andDemulsified Effluent

raw postoxidative effluent demulsified effluent

compound

C1a (mg/m3;n = 3)a

(n = 3) sb (mg/m3)

recalcd headspace concnexpressed as mole fraction

(ppm)

C1a (mg/m3;n = 3)a

(n = 3) sb (mg/m3)

recalcd headspace concnexpressed as mole fraction

(ppm) reduction (%)

pyridine 16.22 0.68 4.96 5.21 0.17 1.59 67.92-methylpyridine 23.12 0.93 6.01 6.12 0.25 1.59 73.52,4-dimethylpyridine 29.87 1.10 6.76 5.20 0.20 1.18 82.6sum of unidentified VNCs(as pyridine)

89.12 5.81 27.24 13.10 0.84 4.00 85.3

aC1 = average headspace concentration. bs = standard deviation.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141512

individual VNCs by 68−85% results in a much lighter load oforganic contaminants passed with the wastewater to atreatment plant. This lowers the cost and time required toeffectively treat wastewater and also lowers biotoxicity of theeffluent toward the activated sludge.3.4. Changes in Total VOC Content in the Effluent

Headspace. The developed procedure enables a rapid andaccurate determination of changes in the VOC content in theeffluents. Chromatographic conditions used ensure completeelution of volatile organic compounds in several minutes.Chemical compounds present in a sample are eluted from thechromatographic column as one peak. The investigationscarried out during validation of the procedure demonstratethat such a solution, despite its simplicity, ensures accurateresults. When separation conditions are changed, an identicalchromatographic system can be used to examine the boilingpoint distribution of sample components using empty columngas chromatography (EC-GC).36,37 Changes in total VOCcontent in the effluent headspace resulting from the removal oforganic phase can be treated as a general measure of theeffectiveness of reduction of VOC content in the effluents. Thetotal VOC content in the headspace of the investigatedbituminous materials is listed in Table 6. The changes of CODand BOD as well as toxicity of the effluents are listed in Table7.

The changes in total VOC content in postoxidative effluentsresulting from the removal of organic phase were comparable tothe change in VNC content. Both for VNCs and for VOCs thereduction in content increases with the hydrophobic nature ofindividual chemical compounds, as their content in rawpostoxidative effluents is determined by the organic phase−aqueous phase partition coefficient.To supplement the above results of changes in VOC content

in the effluents resulting from the removal of organic, the

changes in other quality parameters describing the load andnature of pollutants present in postoxidative effluents arediscussed below.The above results reveal that a technically simple effluent

demulsification results in a substantial reduction in pollutantcontent. BOD and COD values are reduced by more than 60%.The reduction in total VOC content in the effluents is of asimilar magnitude. For strongly hydrophobic compounds thisdegree of reduction is close to 100% in the majority of cases.The developed procedures of determination of the effluentheadspace composition allow a substantial extension ofwastewater quality parameters to include the VOC content.Universal applicability to wastewater treatment processes is anadvantage of the developed procedures. Identical procedurescan be successfully used in the investigation of effectiveness andtransformations taking place during advanced processes ofeffluent oxidation.

4. CONCLUSIONAlkaline effluents from the bitumen production are charac-terized by a complex composition. In their raw form theeffluents are strongly emulsified. The presence of an emulsifiedorganic phase in the aqueous phase of the effluents results intheir high toxicity toward the activated sludge of a wastewatertreatment plant as well as malodorousness. One hundred andfourteen volatile organic compounds were identified in the rawpostoxidative effluents. Analysis of the results revealed that over400 volatile organic compounds were present in the effluentheadspace. If chromatographic resolution were improved, e.g.,by using a two-dimensional separation or preliminaryfractionation, this number would likely be greater.The procedures described in this work enable detailed

identification of VOCs and determination of distribution ofconcentration of individual classes of chemical compounds. Thedetermination of changes in concentration of individual classesof VOCs, including malodorous volatile sulfur compounds, andchanges in total VOC content expands considerably thenumber of data on wastewater treatment processes.The method of preliminary wastewater treatment, involving

the removal of organic phase through effluent demulsification,discussed in this work allows a substantial reduction of the totalpollutant load accompanied by a proportional reduction inVOC content.

■ ASSOCIATED CONTENT*S Supporting InformationFigure showing DHS-GC-MS chromatogram of raw postox-idative effluent. This material is available free of charge via theInternet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: grzegorz.boczkaj@gmail.com. Tel.: (+48) 697970303.Fax: (+48 58) 347-26-94.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank the Lotos Asfalt, Ltd. (Lotos Group, Inc.) for theircooperation on this project. We gratefully acknowledgefinancial support from the National Science Center, Warsaw,

Table 6. Compilation of Total VOC Content in theHeadspace of Bituminous Materials Determined Using SHS-GC-FID

raw postoxidative effluenteffluent after removal of organic

phase

C1a

(n = 3;mg/m3)

sb

(mg/m3)

recalcd head-space concnexpressed asmole fraction

(ppm)

C1a

(n = 3;mg/m3)

sb

(mg/m3)

recalcd head-space concnexpressed asmole fraction

(ppm)reductionin content

5384 205 8227 1786 71 2729 66.8%aC1 = average headspace concentration. bs = standard deviation.

Table 7. Comparison of Changes in Monitored EffluentParameters Resulting from Removal of Organic Phase

raw effluent demulsified effluent

parameterav value(n = 3) sa

v value(n = 3) sa change (%)

COD(mgO2/dm

3)11990 803 4720 325 61

BOD5(mgO2/dm

3)6620 470 2075 145 69

acute toxicity(%)

EC20 0.10 0.20 50EC50 0.40 0.80 50

as = standard deviation.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141513

Poland (Contract nos. N N209 761740 and UMO-2011/01/N/ST8/07757).

■ REFERENCES(1) Paxeus, N. Organic pollutants in the effluents of large wastewatertreatment plants in Sweden. Water. Res. 1996, 30, 1115.(2) Vogelsang, C.; Grung, M.; Jantsch, T. G.; Tollefsen, K. E.;Liltved, H. Occurrence and removal of selected organic micro-pollutants at mechanical, chemical and advanced wastewater treatmentplants in Norway. Water. Res. 2006, 40, 3559.(3) Ratola, N.; Cincinelli, A.; Alves, A.; Katsoyiannis, A. Occurrenceof organic microcontaminants in the wastewater treatment process. Amini review. J. Hazard. Mater. 2012, 239−240, 1.(4) Ras, M. R.; Borrul, F.; Marce, R. M. Determination of volatileorganic sulfur compounds in the air at sewage management areas bythermal desorption and gas chromatography−mass spectrometry.Talanta 2008, 74, 562.(5) Petrasek, A. C.; Kugelman, I. J.; Austern, B. M.; Pressley, T. A.;Winslow, L. A.; Wise, R. H. Fate of Toxic Organic Compounds inWastewater Treatment Plants. J.−Water. Pollut. Control Fed. 1983, 55,1286.(6) Rogers, H. R. Sources behavior and fate of organic contaminantsduring sewage treatment and in sewage sludges. Sci. Total Environ.1996, 185, 3.(7) Schroder, H. Fr. Polar organic pollutants from textile industries inthe wastewater treatment processBiochemical and physicochemicalelimination and degradation monitoring by LC-MS, FIA-MS and MS-MS. TrAC, Trends Anal. Chem. 1996, 15, 349.(8) Cabeza, Y.; Candela, L.; Ronen, D.; Teijon, G. Monitoring theoccurrence of emerging contaminants in treated wastewater andgroundwater between 2008 and 2010. The Baix Llobregat (Barcelona,Spain). J. Hazard. Mater. 2012, 239−240, 32.(9) Miao, X. S.; Koenig, B. G.; Metcalfe, C. D. Analysis of acidicdrugs in the effluents of sewage treatment plants using liquidchromatography−electrospray ionization tandem mass spectrometry.J. Chromatogr. A 2002, 952, 139.(10) Schroder, H. F. Surfactants: Non-biodegradable, significantpollutants in sewage treatment plant effluents: Separation, identi-fication and quantification by liquid chromatography, flow-injectionanalysis−mass spectrometry and tandem mass spectrometry. J.Chromatogr. A 1993, 647, 219.(11) Gomez, M. J.; Gomez-Ramos, M. M.; Aguera, A.; Mezcua, M.;Herrera, S.; Fernandez-Alba, A. R. A new gas chromatography/massspectrometry method for the simultaneous analysis of target and non-target organic contaminants in waters. J. Chromatogr. A 2009, 1216,4071.(12) Castillo, M.; Barcelo, D. Identification of Polar Toxicants inIndustrial Wastewaters Using Toxicity-Based Fractionation withLiquid Chromatography/Mass Spectrometry. Anal. Chem. 1999, 71,3769.(13) Moeder, M.; Schrader, S.; Winkler, M.; Popp, P. Solid-phasemicroextraction−gas chromatography−mass spectrometry of bio-logically active substances in water samples. J. Chromatogr. A 2000,873, 95−106.(14) Simo, R. Trace chromatographic analysis of dimethyl sulfoxideand related methylated sulfur compounds in natural waters. J.Chromatogr. A 1998, 807, 151.(15) Sola, I.; Ausio, X.; Simo, R.; Grimalt, J. O.; Ginebreda, A.Quantitation of volatile sulphur compounds In polluted waters. J.Chromatogr. A 1997, 778, 329.(16) Lestremau, F.; Desauziers, V.; Fanlo, J. L. Headspace SPMEfollowed by GC/PFPD for the analysis of malodorous sulfurcompounds in liquid industrial effluents. Anal. Bioanal. Chem. 2004,378, 190.(17) Boczkaj, G.; Kaminski, M.; Przyjazny, A. Process Control andInvestigation of Oxidation Kinetics of Postoxidative Effluents UsingGas Chromatography with Pulsed Flame Photometric Detection (GC-PFPD). Ind. Eng. Chem. Res. 2010, 49, 12654.

(18) Yan, X. Sulfur and nitrogen chemiluminescence detection in gaschromatographic analysis. J. Chromatogr. A 2002, 976, 3.(19) Grebel, J. E.; Suffet, I. H. Nitrogen−phosphorus detection andnitrogen chemiluminescence detection of volatile nitrosamines inwater matrices: Optimization and performance comparison. J.Chromatogr. A 2007, 1175, 141.(20) Jurado-Sanchez, B.; Ballesteros, E.; Gallego, M. Evaluation ofstationary phases and gas chromatographic detectors for determinationof amines in water. J. Sep. Sci. 2010, 33, 3365.(21) Scully, F. E., Jr; Hartman, A. C. Disinfection Interference inWastewaters by Natural Organic Nitrogen Compounds. Environ. Sci.Technol. 1996, 30, 1465.(22) Van Stee, L. L. P.; Brinkman, U. A. Th. Developments in theapplication of gas chromatography with atomic emission (plus massspectrometric) detection. J. Chromatogr. A 2008, 1186, 109.(23) Van Stee, L. L. P.; Leonards, P. E. G.; Vreuls, R. J. J.; Brinkman,U. A. Th. Identification of non-target compounds using gaschromatography with simultaneous atomic emission and massspectrometric detection (GC-AED/MS): Analysis of municipalwastewater. Analyst 1999, 124, 1547.(24) Boczkaj, G.; Jaszczołt, M.; Przyjazny, A.; Kaminski, M.Application of normal phase high performance liquid chromatographyfollowed by gas chromatography for analytics of diesel fuel additives.Anal. Bioanal. Chem. 2013, 405, 6095.(25) Suschka, J.; Mrowiec, B.; Kuszmider, G. Volatile organiccompounds (VOC) at some sewage treatment plants in Poland.Water.Sci. Technol. 1996, 33, 273.(26) Escalas, A.; Guadayol, J. M.; Cortina, M.; Rivera, J.; Caixach, J.Time and space patterns of volatile organic compounds in a sewagetreatment plant. Water. Res. 2003, 37, 3913.(27) Leach, J.; Blanch, A.; Bianchi, A. C. Volatile organic compoundsin an urban airborne environment adjacent to a municipal incinerator,waste collection centre and sewage treatment plant. Atmos. Environ.1999, 33, 4309.(28) Onkal-Engin, G.; Demir, I.; Engin, S. N. Determination of therelationship between sewage odour and BOD by neural networks.Environ. Modell. Software 2005, 20, 843.(29) Gostelow, P.; Parsons, S. A.; Stuetz, R. M. Odour measurementsfor sewage treatment works. Water. Res. 2001, 35, 579.(30) Stuetz, R. M.; Fenner, R. A.; Engin, G. Characterization ofwastewater using an electronic nose. Water. Res. 1999, 33, 442.(31) Schroder, H. F. Selective determination of non-biodegradablepolar, organic pollutants in waste water related to functional groupsusing flow injection combined with tandem mass spectrometry. Water.Sci. Technol. 1996, 34, 21.(32) Schroder, H. F. Characterization and monitoring of persistenttoxic organics in the aquatic environment. Water. Sci. Technol. 1998,38, 151−158.(33) Schroder, H. F. Substance-specific detection and pursuit of non-eliminable compounds during biological treatment of waste water fromthe pharmaceutical industry. Waste Manage. 1999, 19, 111.(34) Tartakovsky, B.; Lishman, L. A.; Legge, R. L. Application ofmulti-wavelength fluorometry for monitoring wastewater treatmentprocess dynamics. Water. Res. 1996, 30, 2941.(35) Gilgenast, E.; Boczkaj, G.; Przyjazny, A.; Kaminski, M. Samplepreparation procedure for the determination of polycyclic aromatichydrocarbons in petroleum vacuum residue and bitumen. Anal.Bioanal. Chem. 2011, 401, 1059.(36) Boczkaj, G.; Przyjazny, A.; Kaminski, M. New procedure for thedetermination of distillation temperature distribution of high-boilingpetroleum products and fractions. Anal. Bioanal. Chem. 2011, 399,3253.(37) Boczkaj, G.; Kamin ski, M. Research on the separation propertiesof empty-column gas chromatography (EC-GC) and conditions forsimulated distillation (SIMDIS). Anal. Bioanal. Chem. 2013, 405, 8377.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie402126d | Ind. Eng. Chem. Res. 2014, 53, 1503−15141514